Well, you had a intermittent, mechanically-induced plate failure. Probably one of the worst (and rare?) conditions for a battery. Pretty wild. Must have had some bad Karma or something going on
actually not that rare.. we coined that issue YSDS back in the C10 days...
(Yuasa Sudden Death Syndrome) as the older batteries were very poorly manufactured with respect to the inter cell tie bars. Which (as in the case mentioned) would fracture, and offer odd voltages when a draw is placed on the battery... they are always in multples of approx 2 volts each, as each "cell" ( noted by the filler holes, of which there are 6) shows where the mini fracture actually is in the stack, 2 volts, would be a break between the first and second set of plates, 6v would be the center set, that 3.6v tested batt was broken between the second and third set of plates.. and so on... other than just being a very old and sulfated battery, or one that was allowed to dip low on charge, and freeze (which creates shorts between plates, and again shows low) it's the most common... usually it is appearant right off when a load test is done, but testing a cool battery vs a hot one (like during a ride) will exhibit differences as the tie bar metal 'fracture' may, due to expansion, be closed, or, even more open, and only visible after 'bouncing it' during the load test. Construction and materials has gotten exponentially better in the last decade, but there are still a lot of "low end" batteries in service, that simply never see bouncing and such, to make them finally die...
I also think shipped batteries, when you but them singly online, get subjected to some extreme torture rides in transit, unless they are packed robustly. I've had a couple show up looking like the UPS/USPS people played basket ball with the box... and they were bad from the start.